William Liu · Podcasts

All podcasts · Series

Agentic Coding Capability: From Coding Models to Coding Agents

The series overview lays out the shared mental models and expert disagreements that recur throughout the podcast.

59 episodes 59 with audio
Apple Podcasts Spotify Overcast Pocket Casts Android RSS
  1. SE0 — Series Overview — The New Data Unit for Coding Agents
  2. T1E0 — T1E0 · Topic 1 Overview — Foundations of Agentic Coding
  3. T1E1 — T1E1 · From Code Generation to Agentic Work
  4. T1E2 — T1E2 · The Agent-Computer Interface
  5. T1E3 — T1E3 · Trajectories as Training Data
  6. T1E4 — T1E4 · SWE-bench — Real Issues, Real Repositories
  7. T1E5 — T1E5 · SWE-agent and the Agent-Computer Interface
  8. T1E6 — T1E6 · Aider Benchmark — Can the Model Fix Its Own Mistakes?
  9. T1E7 — T1E7 · GitHub Copilot Cloud Agent — From Session to Reviewable Work
  10. T2E0 — T2E0 · Evaluation and Code Review for Coding Agents
  11. T2E1 — T2E1 · How to Evaluate a Coding Agent in Practice
  12. T2E2 — T2E2 · Public Tests, Hidden Tests, and Oracle Tests
  13. T2E3 — T2E3 · Leaderboard Literacy
  14. T2E4 — T2E4 · Code Review Agents: The Other Half of Agentic Coding
  15. T2E5 — T2E5 · Evaluating Review Quality
  16. T2E6 — T2E6 · Code Review Benchmarks and Datasets
  17. T2E7 — T2E7 · SWE-bench Verified
  18. T2E8 — T2E8 · SWE-bench-Live
  19. T2E9 — T2E9 · Aider Polyglot Benchmark
  20. T2E10 — T2E10 · Terminal-Bench 2.0
  21. T2E11 — T2E11 · FeatureBench
  22. T2E12 — T2E12 · CodeClash
  23. T2E13 — T2E13 · LLM-as-a-Verifier
  24. T2E15 — T2E15 · CodeReviewer
  25. T2E16 — T2E16 · CodeReviewQA
  26. T2E17 — T2E17 · CodeFuse-CR-Bench
  27. T2E18 — T2E18 · CR-Bench: Evaluating the Real-World Utility of AI Code Review Agents
  28. T2E19 — T2E19 · SWE-PRBench
  29. T2E20 — T2E20 · CodeReviewBench: Planted Bugs, Two Judges
  30. T2E21 — T2E21 · Martian's Code Review Bench: Graded by Real Developer Behavior
  31. T2E22 — T2E22 · MetaMateCR: AI-Assisted Fixes to Code Review Comments at Scale
  32. T3E0 — T3E0 · Agent Systems, Harnesses, and Observability
  33. T3E1 — T3E1 · GitHub Copilot Coding Agent: From Task to Reviewable Work
  34. T3E2 — T3E2 · OpenAI Codex: Tasks, Workflows, and the Agent-First Loop
  35. T3E3 — T3E3 · Building a Multi-Agent Research System
  36. T3E4 — T3E4 · Effective Harnesses for Long-Running Agents
  37. T3E5 — T3E5 · Harness Design for Long-Running Application Development
  38. T3E6 — T3E6 · Harness Engineering in an Agent-First World
  39. T3E7 — T3E7 · Meta-Harness: End-to-End Optimization of Model Harnesses
  40. T3E8 — T3E8 · The Harness Problem
  41. T3E9 — T3E9 · Harness Engineering as a Discipline
  42. T4E0 — T4E0 · Training Signals and Improvement Studies
  43. T4E1 — T4E1 · CWM — Code Generation with World Models
  44. T4E2 — T4E2 · Debugging Code World Models
  45. T4E3 — T4E3 · SWE-Gym — Training Agents and Verifiers
  46. T4E4 — T4E4 · SWE-smith — Scaling Data for SWE Agents
  47. T4E5 — T4E5 · SWE-RL — Reinforcement Learning on Software Evolution
  48. T4E6 — T4E6 · Agent-RLVR — Guidance and Environment Rewards
  49. T4E7 — T4E7 · R2E-Gym — Procedural Environments and Hybrid Verifiers
  50. T4E8 — T4E8 · SWE-Hub — A Production System for Executable SWE Tasks
  51. T4E9 — T4E9 · Nemotron-Terminal — Data Engineering for Terminal Capability
  52. T4E10 — T4E10 · SWE-Bench Mobile
  53. T5E0 — T5E0 · Data Products for Agentic Coding Training
  54. T5E1 — T5E1 · The Agentic Coding Data Record
  55. T5E2 — T5E2 · Trajectory Data for Training and Analysis
  56. T5E3 — T5E3 · Verification and Review Labels
  57. T5E4 — T5E4 · From Production Telemetry to Training Data
  58. T5E5 — T5E5 · Governance and Splits
  59. T5E6 — T5E6 · The Six-Layer Content Schema